32 Mr. W. J. Hamilton and Mr. H. E. Strickland, on the Geology of 



a circuit round the north extremity of an overhanging cliiF of lava nearly 50 

 feet high. The lava is vesicular and scoriaceous in the upper part, for about 6 

 or 7 feet. The next 20 feet are irregularly columnar, which structure gradually 

 disappears in the lower part of the bed. The marls and limestones on the opposite 

 or N.E. side of the river have been here much altered by their contact with the 

 igneous rock, having been converted, in many places, into a yellow jasper-like 

 substance, with a shining lustre and conchoidal fracture. 



This lava encircles the base of several cones of the second period, and overlies 

 the older coulees which they have sent forth (PI. II.). In one place, 2^ miles N.N.E. 

 from Karadewit, a narrow stream of lava has diverged to the east from the general 

 mass, through a lateral valley in the schistose range, and has spread over the 

 coulee from the easternmost cones of the second period before-described. The 

 contrast between the lava-streams of the two periods is most striking. 



The second of these recent volcanic cones rises in the plain between Sandal and 

 Megne*, and is also called Karadewit (PL II, and PI. III., Sec. 7). This hill consists 

 of scorise and ashes resting on a more solid crust of red scoriaceous rock. The crater 

 is quite perfect, and from 150 to 200 feet deep, the bottom being choked with large 

 stones which have rolled down its sides. Many considerable caverns and hollows 

 exist in the scoriaceous rock. The summit consists of a very narrow ridge surround- 

 ing the crater. A few small pine-trees grow on the sides of this cone, which, on the 

 whole, bears more vegetation than the Karadewit of Koola. A stream of lava has 

 issued from the west foot of the cone, and flowed northwards for about five miles 

 to the Hermus ; and a small crater, about a mile to the west of the great one, has 

 sent forth a narrow coulee which soon joins the general mass. There is also a low 

 flat cone to the south-east, apparently belonging to the 2nd period, from the crater 

 of which a small stream of lava, about half a mile in length, has been poured 

 forth, apparently at the same time as the eruption from the large cone. This lava 

 is remarkable from having flowed over the margin of the crater, instead of bursting 

 through the sides, like all the other lava streams in the district. This is probably 

 owing to its having flowed from the crater of an older and therefore more consoli- 

 dated cone ; whereas the recent cones, being formed of loose scorise and ashes, and 

 much higher, were unable to support the mass of heated lava until it should over- 

 flow at the summit, and therefore yielded at their bases a passage for the lava. 



The third or most western of the three great cones is called by the Turks Ka- 

 plan Alan (PI. II. and PI. III., Sec. 7). It is surrounded on all sides for more than 

 a mile by the black and rugged lavas, which have flowed from its base, and render 



* Near the centre of the plain of Megne a well has been sunk through the alluvial soil to the mica- 

 ceous schist below. 



